CN108540913A - Make the method that audio signal frequency is distorted and the hearing devices to work according to this method - Google Patents
Make the method that audio signal frequency is distorted and the hearing devices to work according to this method Download PDFInfo
- Publication number
- CN108540913A CN108540913A CN201810181968.7A CN201810181968A CN108540913A CN 108540913 A CN108540913 A CN 108540913A CN 201810181968 A CN201810181968 A CN 201810181968A CN 108540913 A CN108540913 A CN 108540913A
- Authority
- CN
- China
- Prior art keywords
- signal
- frequency
- low
- component
- components
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
- H04R25/00—Deaf-aid sets, i.e. electro-acoustic or electro-mechanical hearing aids; Electric tinnitus maskers providing an auditory perception
- H04R25/45—Prevention of acoustic reaction, i.e. acoustic oscillatory feedback
- H04R25/453—Prevention of acoustic reaction, i.e. acoustic oscillatory feedback electronically
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
- H04R25/00—Deaf-aid sets, i.e. electro-acoustic or electro-mechanical hearing aids; Electric tinnitus maskers providing an auditory perception
- H04R25/35—Deaf-aid sets, i.e. electro-acoustic or electro-mechanical hearing aids; Electric tinnitus maskers providing an auditory perception using translation techniques
- H04R25/353—Frequency, e.g. frequency shift or compression
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
- H04R25/00—Deaf-aid sets, i.e. electro-acoustic or electro-mechanical hearing aids; Electric tinnitus maskers providing an auditory perception
- H04R25/50—Customised settings for obtaining desired overall acoustical characteristics
- H04R25/505—Customised settings for obtaining desired overall acoustical characteristics using digital signal processing
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
- H04R2225/00—Details of deaf aids covered by H04R25/00, not provided for in any of its subgroups
- H04R2225/43—Signal processing in hearing aids to enhance the speech intelligibility
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
- H04R2430/00—Signal processing covered by H04R, not provided for in its groups
- H04R2430/03—Synergistic effects of band splitting and sub-band processing
Abstract
The present invention relates to a kind of methods for making audio signal frequency be distorted and the hearing devices to be worked according to this method.It gives for making as input signal (E existing for audio signalk) frequency distortion method and relevant hearing devices (2).By input signal (Ek) it is divided into low-frequency signal components (NF) and high frequency component signal (HF), wherein the two signal components (NF, HF) are in edge frequency (fg) at have a common boundary each other.Make high frequency component signal frequency distortion, and be superimposed with low-frequency signal components, to form output signal (A).At least for including edge frequency (fg) high frequency component signal (HF) and/or low-frequency signal components (NF) fringe region (RH, RL), change associated amplification factor so that the signal level (L of low-frequency signal components (NF)1) and frequency distortion high frequency component signal (HF) signal level (L2) between level error (Δ L) increase.
Description
Technical field
The present invention relates to a kind of methods for making audio signal frequency be distorted.The invention further relates to a kind of according to the party
The hearing devices of method work.
Background technology
Following device is generally known as " hearing devices ", which will be feed-in or by record ambient sound generation
Audio signal (hereinafter referred to as " input signal ") is believed in a manner of amplifying and/or being modulated in other ways as sound
Number using user it is appreciable in the form of (such as the air-borne sound or solid sound being fed in duct) output.In addition to wear-type ear
Except machine, especially auditory prosthesis also belongs to hearing devices.It generally again will be for improving the ambient sound to being applied to user's ear
The portable hearing devices of the perception of sound are known as " auditory prosthesis ".The commonly known as subclass quilt of the auditory prosthesis of " hearing aid "
It is configured for supplying on medical significance by the hearing loss person that listened to barrier.
In order to meet the numerous demands of individuals of user, the auditory prosthesis of different structure form is provided, such as worm behind ear helps
Device for tone frequencies (HdO), ((receiver is in duct by RIC, Receiver in the Canal for the auditory prosthesis with external earpiece
In)), inner ear type auditory prosthesis (IdO) or also external ear formula auditory prosthesis or duct auditory prosthesis (ITE, CIC).It is exemplary
The auditory prosthesis that ground is listed is worn on external ear or in duct.In addition, in the market also have bone conduction type hearing aid, implantable or
Vibrating tactile formula hearing aid.Impaired hearing is stimulated in the form of mechanically or electrically in such devices.
Recently, it other than typical hearing aid described above, is also useful for assisting the hearing aid dress of the people of normal good hearing
It sets.This auditory prosthesis is also referred to as " personal voice amplification product " or " personal voice multiplying arrangement " and (is abbreviated as：
“PSAD”).These PSAD are used to improve the hearing of normal person, and it is most of dedicated for specific hearing situation (such as with
In improving the perception to animal sounds, improves dialogue in complicated noise circumstance and understand or for pointedly inhibiting environment
Noise).
In the hearing devices of the above-mentioned type, often with frequency distortion, especially frequency shift (FS) and/or frequency compression
Mode reproduces the input signal of feed-in.Herein, on the one hand, often frequency of use is distorted during feedback inhibition, and just
Make it possible to preferably estimated feedback signal for this, so that can preferably carry out feedback inhibition and reduce reproduction
Signal in pseudo- sound.On the other hand, often frequency of use is distorted in hearing aid, will listen to barrier will pass through person is often special
The high frequency noise components that can poorly perceive are mapped to lower frequency, enabling the person that listens to barrier is to sound (especially language for improvement
Sound) perception.
However, in both cases, frequency distortion is usually not applied to entire tone spectrum, and is only applied to tone spectrum
It is more than the high frequency component signal of predetermined margin frequency.
The hearing device of the method for preamble according to claim 1 and preamble according to claim 7 is from EP
It is known in 2 244 491 B2.Here, dividing input signals into high frequency component signal and low frequency signal point by dividing network
Amount, wherein make high frequency component signal frequency distortion.Then, the high frequency component signal of low-frequency signal components and frequency distortion is folded
Add, to form output signal.The theme of 2 244 491 B2 of document EP is following problem：Since practical dividing network is inaccurate
Really, two signal components always have certain spectrum overlapping in the range of edge frequency.It is well known that due to this heavy
Folded, frequency distortion may lead to feature puppet sound, (that is, frequency when especially having dominant frequency in overlapping region when input signal
Spectrum peak, especially loud sine wave (Sinuston)).That is, in this case, making with high-frequency signal point
A part of frequency distortion of the dominant frequency of amount, and another part holding of the dominant frequency with low-frequency signal components is not lost
Very.Therefore the dominant frequency of input signal is mapped in two close adjacent frequencies of output signal, cause audible
And the beat (Schwebung) through being felt frequently as interference.According to 2 244 491 B2 of EP, by offset boundary frequency,
So that the pseudo- sound in output signal reduces, to mitigate the problem.
A kind of hearing devices are it is known that from 00/02418 A1 of WO, are divided into input signal by dividing network
In low frequency and high frequency band, and the amplitude of the signal of two frequency bands is adjusted by two AGC.By controlling signal adjustment
The compression ratio of AGC, wherein the raising of a compression ratio makes another compression ratio while reducing.It then, will using adder
The frequency band superposition of two amplifications
Finally, a kind of known method for inhibiting the acoustic feedback in auditory prosthesis from 2 988 529 A1 of EP.
In the method, the frequency range that will be transmitted by auditory prosthesis is divided into two frequency ranges by separating.In a frequency
The transmission function of feedback path is estimated within the scope of rate and analyzes its behavior at frequency dividing.According to analysis as a result, reduce or
Frequency dividing is improved, and applies phase and/or frequency shift in upper frequency range, to carry out feedback inhibition.
Invention content
The technical problem to be solved by the present invention is to provide a kind of method for making audio signal frequency be distorted, utilize this
Method can particularly effectively inhibit the pseudo- sound of previously described type.In addition, the technical problem to be solved by the present invention is to carry
For a kind of hearing devices for the pseudo- sound for particularly effectively inhibiting previously described type.
According to the present invention, above-mentioned technical problem is solved by the method for the feature with the present invention.According to the present invention,
Above-mentioned technical problem is also solved by the hearing devices of the feature with the present invention.Part sheet is provided in the following description
Body is considered as creative advantageous implementation and extension.
It is used to that audio signal frequency be made to be distorted according to the method for the present invention, especially when hearing devices are run.It will be following
The audio signal of referred to as " input signal " is divided into low-frequency signal components (hereinafter referred to as：" NF components ") and high-frequency signal point
Amount is (hereinafter referred to as：" HF components ").The frequency that the two signal components are had a common boundary each other hereinafter referred to as " edge frequency ".Term
" low-frequency signal components " (" NF components ") and high frequency component signal (" HF components ") are herein only in the frequency spectrum weight of high frequency component signal
In the sense that the heart is in compared to the barycenter of frequency spectrum of low-frequency signal components at higher frequency, indicate that these signal components are opposite
In mutual spectrum position.
Preferably, the frequency spectrum of input signal is completely covered in NF components and HF components.Therefore, in this case, input letter
Number only it is divided into two mentioned signal components.However, in principle, within the scope of the invention, in addition to NF components and HF points
Except amount, other signals component can also be exported from input signal, be located in tone spectrum on HF components and/or NF divides
Under amount, and distinguished respectively by the type of frequency distortion and adjacent signal component.
According to this method, HF component frequencies are made to be distorted, especially frequency shift (FS) or compression.Term " frequency shift (FS) " is herein
The HF components of input signal are mapped on another spectral regions of same spectral range by expression.On the contrary, term " compression " table
Show and HF components are mapped on the spectral regions of smaller spectral range.In principle, frequency distortion is within the scope of this invention
Can exist alternatively in the form of " extension ", i.e., is mapped to HF components on the spectral regions of the spectral range of bigger, even if
This frequency distortion is not common in hearing devices at present.
It is preferred that being distorted NF component frequencies, therefore remained unchanged in terms of its spectrum position and range.However, and this
Difference, within the scope of this invention, can also to NF components carry out frequency distortion, however its in this case with HF components
Frequency distortion form differently.
According to this method, NF components are superimposed with the HF components of frequency distortion, to form output signal.
Here, optionally, to before the frequency partition of NF components and HF components, or in frequency partition and NF components and
Between the superposition of the HF components of frequency distortion (and herein selectively before or after frequency distortion), also input is believed
Number carry out one or more other signal processing steps, such as analog-digital conversion, the amplification with frequency dependence, feedback suppression
System etc..Similarly, output signal can be carried out within the scope of the invention other signal processing (such as digital-to-analog turn
It changes and/or amplifies).
According to the present invention, the frequency spectrum fringe region at least for HF components and/or NF components including edge frequency changes
Become, increase or reduce associated amplification factor so that the signal of the signal level of NF components and the HF components of frequency distortion
Level error between level increases.If the change of amplification factor is not related to entire NF or HF components, and only relates to its marginal zone
Domain can use the signal level in the fringe region then when determining level error.Particularly, by the NF at dominant frequency points
The signal level of amount and HF components is compared to each other, to determine level error.Advantageously it is amplified the change of the factor so that eliminate
Or the audible beat at least in the overlapping region of reduction HF components and NF components.
The present invention is based on following understanding：NF components and the dominant frequency of the input signal in the HF components of frequency distortion
Signal level is more similar, then can more significantly perceive the pseudo- sound of beginning description.By according to the present invention under any circumstance
Increase the level error between NF components and HF components in the fringe region of NF components and HF components, it is known that especially effectively
Ground reduces the perception to pseudo- sound.
In principle, within the scope of the invention it is contemplated that as described in 2 244 491B2 of EP, such as by means of
Help dividing network that input signal is accurately divided into two (not going on subdivision) signal components in itself, i.e., NF components and
HF components.However, in the preferred implementation of the present invention, input signal is divided using filter group, filter group believes input
It number is divided into multiple (that is, far more than two, but at least four) in frequency band.In the general construction of this filter group
In, input signal is for example divided into 48 frequency bands.
According to this method, thus multiple high frequency bands conduct HF components.Correspondingly, make these in the manner described before
High frequency band frequency distortion.Conversely, multiple low-frequency bands conduct NF components.Correspondingly, these frequency bands not frequency distortion or opposite
In HF components frequency distortion in different ways.Term " high frequency " (" HF ") and " low frequency " (" NF ") are interpreted as opposite again herein
Instruction.In addition, in the sense that explanation above, there may be frequencies higher than " high frequency " frequency band or less than low-frequency band
Other frequency bands, neither with HF components nor associated with NF components, but in contrast since different types of frequency is lost
Very protruded as other signals component.
The fringe region of high frequency component signal is herein when needed by the subset shape for the high frequency band having a common boundary with low-frequency band
At.It as its supplement or replaces, the fringe region of low-frequency signal components is by the subset shape for the low-frequency band having a common boundary with high frequency band
At.
Term " subset of frequency band " represents less than the frequency band of the quantity of the sum of the frequency band of correlated signal components herein, and
And in the limiting case can also only include single frequency band.In fact, the corresponding fringe region of HF components or NF components is by list
This limiting case that a frequency band is formed, which is that one of the present invention is preferred, to be implemented.In the sense that, plural "(frequency band) " is interpreted as the case where including single frequency band.
Corresponding fringe region and frequency band associated there are characterized in that, are different from remaining frequency of HF or NF components
Band only changes amplification factor in the frequency band of corresponding fringe region, with the signal relative to correspondingly another signal component
Level increases level error.
Especially select the fringe region of NF components and/or HF components so that its spectral range includes NF components and HF points
The spectrum overlapping region of amount.If input signal is divided on multiple frequency bands, corresponding fringe region is especially by including
The frequency band of overlapping region is formed.
In the advantageous implementation of the present invention, one in only two signal components (is only HF components or only NF
Component) fringe region that defines to increase level error and change amplification factor, simultaneously amplify the factor correspondingly another believe
It is kept constant in number component.However, the present invention a particularly advantageous implementation in, unlike this, not only NF components,
And fringe region is defined respectively for HF components.Amplification factor in the two fringe regions changes with being always mutually opposite herein
Become.As a result, in the fringe region of the first signal component in two signal components (i.e. HF components or NF components), increase is put
The big factor, while reducing the amplification factor in the fringe region of second signal component (i.e. NF components or HF components).
In the particularly advantageous deformation of the present invention, reduce the amplification factor in second signal component at this so that
Thereby compensate for the increase of the amplification factor in the first signal component.Therefore, in other words, reciprocally change two edges
Amplification factor in region so that letter average on average signal level or two fringe regions on two fringe regions
Number power constant (that is, not influenced by amplification factor variation).This is especially in input signal in HF components and NF components
Overlapping region in tone property it is strong in the case of (that is, in the overlapping region exist very dominant frequency the case where
Under), cause the change according to the present invention as amplification factor that cannot be perceived in the output signal or only can be very small
Degree on be perceived, especially, the appreciable volume of dominant frequency is not influenced or only non-by horizontal change by level changes
Often small influence.
The change of amplification factor is so that the pseudo- sound of frequency distortion is obviously reduced or even is eliminated, and in terms of it
It will not have a negative impact to the quality reproduction of input signal.Specifically, the sine wave near edge frequency with almost with
Identical volume is reproduced in conventional method, however wherein, usually the beat quilt of these sine waves caused by frequency distortion
It completely eliminates or is at least largely eliminated.
In the advantageous extension of the present invention, the increase of level error according to the present invention does not carry out unconditionally, and
It is that only when this is really significant (or on this really significant degree), i.e., ought be expected to occur to hear in the output signal
Pseudo- sound (or accordingly expected intensity for expected pseudo- sound occur) when carry out.It is well known that if input signal is at HF points
There is high tone in the spectrum overlapping region of amount and NF components, i.e., it is (special if there are dominant frequencies in the overlapping region
It is loud sine wave), then it is expected audible pseudo- sound occur.Therefore, in this extension of the method, acquisition characterization
The feature of the tone of input signal in overlapping region (i.e. in other words, forms the tone of the input signal in overlapping region
Estimated value or reduced value) characteristic parameter.
According to this method, the change of amplification factor according to the present invention and the therefore water between HF components and NF components
The increase of adjustment is carried out according to this feature parameter herein.Particularly, here, only when this feature parameter meets preassigned, especially
It is when being more than predetermined threshold, just to increase level error.In the replacement embodiment of the present invention, according to this feature parameter (line
Property or non-linearly) increase of level error is weighted.Characterize the spy of the feature of the tone of the input signal in overlapping region
Sign parameter is preferably determined by the auto-correlation of the input signal in overlapping region at this.Particularly, characteristic parameter by (
In mathematical meaning for complex values) magnitude of auto-correlation function formed.
Hearing devices according to the present invention are typically configured as executing previously described side according to the present invention automatically
Method.The embodiment of previously described method and extension correspondingly correspond to related embodiment and the extension of device, wherein
The advantages of these methods deform can also be diverted to the corresponding embodiment of hearing devices.Specifically, according to the present invention
Hearing devices include frequency divider, and signal will be received by, which being configured as, is divided into low-frequency signal components (NF components) and high-frequency signal point
It measures (HF components), wherein the two signal components are had a common boundary each other at edge frequency.In addition, hearing devices include：At signal
Device is managed, is configurable for making high frequency component signal frequency distortion；And synthesizer, it is configurable for believing low frequency
Number component and the superposition of the high frequency component signal of frequency distortion, to form output signal.
According to the present invention, signal processor is configured as, at least for HF components and/or NF point including edge frequency
The frequency spectrum fringe region of amount, changes associated amplification factor so that the signal level of NF components and the HF components of frequency distortion
Signal level between level error increase.
Preferably, frequency divider is formed by (analysis) filter group, which is configured as, and input signal is decomposed
Into multiple frequency bands.Synthesizer is accordingly formed by (synthesis) filter group in this embodiment, then which exists
Frequency band is merged into output signal after frequency distortion (and possible other signals processing step).In addition, at for signal
The implementation deformation for managing device, similarly with reference to the description being previously with regard to according to the method for the present invention.
Hearing devices according to the present invention auditory prosthesis in particular, be further preferably used to the supply person that listens to barrier herein and
The hearing aid of structure.
Description of the drawings
The embodiment of the present invention is explained in detail below in conjunction with the accompanying drawings, is distinguished herein：
Fig. 1 schematically shows the hearing devices of hearing aid styles with block diagram, is filtered by (analysis) in hearing aid
The audio signal (input signal) of input is divided into multiple frequency bands by device group, wherein the input signal conducted in each frequency band
Low-frequency signal components (NF components) and high frequency component signal (HF components) are subdivided at edge frequency, wherein pass through signal
Processor makes the HF component frequencies of input signal be distorted, and wherein, by the HF of frequency distortion points in (synthesis) filter group
Amount is superimposed with the NF components of input signal,
Fig. 2 schematically shows the amplitude of (analysis) filter group with the curve of the signal gain relative to frequency
Frequency response,
Fig. 3 schematically shows the side for making frequency input signal be distorted executed by hearing devices with flow
Method, and
Fig. 4 and Fig. 5 is respectively schematically with the curve graph of the signal gain relative to frequency, for two kinds of variety classes
Input signal, by directly at edge frequency have a common boundary two frequency bands amplitude frequency response, show the method
Effect.
In all the appended drawings, identical reference numeral is arranged to the part to correspond to each other and parameter respectively.
Specific implementation mode
Fig. 1 shows the hearing devices of 2 form of hearing aid.Hearing aid 2 includes input converter 4, subtracts as the basic element of character
Musical instruments used in a Buddhist or Taoist mass 6, (analysis) filter group 8, signal processor 10, (synthesis) filter group 12, output translator 14 and with arrangement
The electric feedback path 16 of (adaptive) filter 18 wherein.
The sound that (illustratively being formed in the current situation by microphone) input converter 4 will be inputted from ambient enviroment
Sound signal SiBe converted to (initial) input signal Ei。
In order to inhibit acoustic feedback, from original input signal E in subtracter 6iIn subtract and produced in electric feedback path 16
Raw electronic compensating signal K.By input signal Ei(having been compensated for) input signal E is generated with the subtraction of thermal compensation signal Kk, conveyed
To (analysis) filter group 8.
In filter group 8, by input signal EkMultiple frequency band F are divided on frequency spectrumjIn.Here, parameter j is to be used for
To frequency band FjCarry out the Numerical Index of serial number.In the simplification example according to Fig. 1 to Fig. 5, filter group 8 believes input
Number EkIt is divided into six frequency band Fj(wherein j=1,2 ..., 6) in, also specifically be indicated as F1To F6.In the reality of hearing aid 2
In the application implementation on border, filter group 8 is by input signal EkIt is divided into much more (for example, 48) frequency channel FjIn.
In signal processor 10, to decomposing frequency band FjIn input signal EkIn a manner of specific to frequency band
Reason.Frequency band F (and is divided on frequency spectrum by treated the signal P of signal processor 10j' (j=1,2 ..., 6) in) quilt
It is transported to (synthesis) filter group 12, by frequency band Fj' it is electrical output signal A to merge (superposition).
On the one hand output signal A is transported to (such as being formed by loud speaker or " receiver ") output translator 14,
Output signal A is converted into output voice signal Sa。
On the other hand output signal A is transported to sef-adapting filter 18 via electric feedback path 16, thereby determine that benefit
Repay signal K.Have been compensated for input signal EkIt is transported to sef-adapting filter 18 as with reference to variable with being attached.
When hearing aid 2 is run, by voice signal SaIt is directly output in the duct of hearing aid wearer, or via
Sound tube is transported in duct.But especially in the implementation that hearing aid 2 is disposed in the hearing aid 2 in duct in itself, output
Voice signal SaA part be inevitably used as feedback signal R via acoustic feedback path 20 (such as via hearing aid
2 venting channels or via solid sound) feed back to input converter 4, feedback signal R is superposed to ambient sound there
The voice signal S of inputi。
Voice signal Si、SaIt is original sound signal, especially air-borne sound and/or solid sound herein with feedback signal R.Phase
Instead, input signal Ei、Ek, treated, and signal P, output signal A and thermal compensation signal K are audio signals, that is, transmit sound letter
The electric signal of breath.
In region between analysis filter group 8 and composite filter group 12, associate audio signal, i.e. input signal
EkWith treated signal P, as described above to be broken down into frequency band F on frequency spectrumjOr Fj' in mode conduct.
The digital deaf-aid in particular of hearing aid 2, wherein the signal processing in signal processor 10 is by digital technology
It carries out.In this case, audio signal is modeled the-digitlization of digital quantizer 22 before signal processing, and in signal
By 24 converted back into electrical analog signal of digital-analog convertor after processing.In shown example, analogue-to-digital converters
22 are connected directly between 8 front of filter group, and therefore act on and have been compensated for input signal Ek, while digital-analog convertor
24 are connected to behind filter group 12.In this case, electric feedback path 16 conducts output signal in the form of analog signal
A and thermal compensation signal K.
As its replacement, analogue-to-digital converters 22 are connected between input converter 4 and subtracter 6, therefore are acted on
In original input signal Ei(not shown).In this case, electric feedback path 16 conducts advantageously in the form of digital signal
Output signal A and thermal compensation signal K.
In another kind (same unshowned) embodiment of hearing aid 2, subtracter 6 is connected to analysis filter group 8
Below.Here, frequency band Fj' or the output signal A decomposed on frequency spectrum by other frequency analysis is transferred to adaptively
Filter 18.Sef-adapting filter 18 includes the channel of respective numbers.
Signal processor 10 is in a manner of being generally used for hearing aid, in frequency band FjThe input signal E of middle transmissionkIt carries out each
Kind signal processing, the amplification especially specific to frequency band changed, so that input signal EiReproduction be adapted to hearing
The demands of individuals of impaired hearing aid user, and so that it can hear as far as possible.In addition, signal processor 10 is held
Line frequency is distorted, by output signal A and input signal EiDecorrelation, to realize improved feedback inhibition.
In order to illustrate the effect of frequency distortion, in fig. 2, (also referred to as with the signal gain g with frequency dependence：Gain)
Curve graph the frequency response of analysis filter group 8 is shown relative to frequency f.Signal gain g can also take less than 1 herein
Value, and in this case so that input signal EkWeaken (decaying).
In fig. 2 it can be seen that (being reduced to six in this example) frequency band FjAmplitude frequency response, frequency band FjQuilt
It is divided into three low-frequency band F1、F2And F3And three high frequency band F4、F5And F6.Low-frequency band F1-F3The letter of conduction input herein
Number EkLow-frequency signal components NF, and high frequency band F4-F6Conduct input signal EkHigh frequency component signal HF.
Frequency band F in addition to being transferred to signal processor 10jExcept, frequency band F is further depicted in fig. 2j', frequency band Fj' conduction
By the frequency band F of the signal P that exports that treated of signal processor 10j', and in frequency band Fj' in reflect by signal processor
10 frequency distortions carried out.As can be seen from Figure 2, high frequency component signal HF, i.e. high frequency are acted only in this frequency distortion
Frequency band F4‘-F6 ‘, mode is these frequency bands F4 ‘-F6 ‘Relative to corresponding initial band F4-F6Distinguish at identical bandwidth micro-
Smallly to high frequency f offsets.In contrast, in the frequency band F of low-frequency signal components NF1-F3On, signal processor 10 not into
Line frequency is distorted, to the frequency band F of treated signal P1 ‘-F3 ‘In terms of its spectrum position with initial band F1-F3Unanimously.
Change each frequency band F relative to each other1 ‘-F6 ‘The signal processing of respective amplification factor, in order to clearly rise
See and do not shown in schematic diagram according to fig. 2, therefore here it is shown that all frequency band F1 ‘-F6 ‘Signal gain g having the same.
Frequency band F1-F6With corresponding frequency band F1‘-F6' bandwidth especially given by half breadth (Halbwertsbreite)
Go out.The half value level is in diagram according to fig. 2 for example corresponding to the baseline of the curve graph (abscissa).
It can also be seen that frequency band F from Fig. 21To F6Therefore signal component NF and HF are also overlapped on frequency spectrum.Signal point
The overlapping region U of NF and HF is measured herein by the corresponding external frequency band F of low-frequency signal components NF or high frequency component signal HF3And F4
The spectral distance on corresponding external half value boundary form (referring to Fig. 2).Frequency band F3And F4Amplitude frequency response curve phase
The center of the overlapping region U of friendship defines the edge frequency f of signal component NF and HF hereing.Two adjacent frequency band F3And F4
Form the fringe region R for the low-frequency signal components NF for correspondingly accommodating overlapping region ULOr the edge of high frequency component signal HF
Region RH。
In order to when hearing aid 2 is run therefore when carrying out frequency distortion according to fig. 2, avoid opening in output signal A
The pseudo- sound of the type of head description, signal processor 10 change adjacent according to the method summarized in figure 3 with illustrative construction
Frequency band F3' and F4' (therefore fringe region RLAnd RH) the associated amplification factor of difference.Therefore, respectively with frequency band F3' and
F4' curve of associated amplitude frequency response changes due to this of associated amplification factor in diagram according to fig. 2
Become, deviates upward or downward to a certain extent, referring to Fig. 4 and Fig. 5.
In the first step 30 of (showing a part for the method for running hearing aid 2) the method, at signal
It manages device 10 and receives input signal Ek, as described above, input signal EkFrequency band F is divided by filter group 8jIn, therefore it is implicit
Ground is also divided into signal component NF and HF.
In step 32 then, signal processor 10 is about adjacent frequency band F3And F4(therefore about signal component NF
With the respective fringe region R of HFLAnd RH) it is respectively formed auto-correlation function, indicate fringe region R to obtainLAnd RHIn it is defeated
Enter signal EkTone quantitative measurement characteristic parameter.
As mentioned above, term " tone " indicates input signal EkCharacteristic, characteristic present's single frequency f0(figure
4 and Fig. 5) by frequency band F3And F4Advantage in the frequency range of covering.If with specific frequency (under the frequency, frequency
The signal level of resolution is significantly more than average signal level) advantage tone (such as violin tone) characterize fringe region
RLAnd RHIn input signal Ek, then high tone is provided herein.In contrast, if adjacent frequency band F3And F4Signal by
Dominant in the noise component(s) (such as noise, traffic noise, voice noise etc.) in broadband, then tone is low.
The method is the knowledge of the good measurement of tone using auto-correlation function herein.Especially in filter group 8
It is the excellent of the present invention of filter group (i.e. the filter group based on discrete Fourier transform) or similar realization that DFT is modulated
It selects in embodiment, frequency band F3And F4In sinusoidal signal correspond to rotation plurality of indicators, under constant frequency even
It is jumped and is rotated with constant angle between continuous time step.At a step auto-correlation (" one-tap-autocorrelation ")
In, such as preferably in the step 32 of the method determined by, by the pointer maps of the rotation to have and the angle step pair
On the plurality of indicators at the constant phase angle answered.
Here, signal processor 10 uses the magnitude of the auto-correlation function of the complex values as the measurement of tone.It replaces
Ground, using the variance at plurality of indicators or phase angle as the measurement of tone, wherein stable frequency is indicated using small variance,
Thereby indicate that high-pitched tone.Signal processor 10 is according to the phase angle of the auto-correlation function of the complex values, export dominant frequency f0's
Magnitude, mode are that the magnitude of the time interval between the phase angle divided by two time steps (is specially by signal processor：), whereinRepresent phase angle, TsRepresent the time interval；Dominant frequency f0Herein relate to corresponding frequency
Band T3Or T4Band center.
In step 34, signal processor 10 is by as shown in FIG. 2, by initial band F4-F6It is inclined to be converted to frequency
Frequency band F after shifting4‘-F6', to carry out frequency distortion.
In step 36, signal processor 10 checks the measurement of predetermined tone, i.e. for example in frequency band F3And F4In
Whether the magnitude of determining auto-correlation function is less than scheduled threshold value.
If situation is "Yes", this is identified as being contemplated by frequency distortion and does not cause interference pseudo- by signal processor 10
The instruction of sound.Correspondingly, signal processor 10 jumps to step 38 in method execution in this case, in step 38,
It is in frequency band Fj' in the signal P of frequency distortion is output to filter (when needed after executing other signal processing step)
Wave device group 12, with synthesized output signal A.
Otherwise, if the measurement that the inspection carried out in step 36 obtains tone is not less than scheduled boundary value ("No"),
Then in step 40, the estimation of signal processor 10 is in dominant frequency f0Place or the dominant frequency f after offset0' the adjacent frequency in place
Band F3' and F4' in level error Δ L (Fig. 4 and Fig. 5).Signal processor 10 particularly determines that level error Δ L, mode are herein,
It is as in frequency f0Or f0' place frequency band F3' and F4' in signal level L1And L2Measurement, determine each of amplitude frequency response
A curve is in these frequencies f0And f0' value at place, and be compared each other (Δ L=| L1–L2|；Referring to Fig. 4 and Fig. 5).
In step 42 then, signal processor 10 checks whether predetermined level error Δ L has been more than scheduled
Boundary value.
If situation is "Yes", this is identified as due to just high level error Δ L, expection originally by signal processor 10
It will not cause the instruction for interfering pseudo- sound due to frequency distortion.Correspondingly, signal processor 10 is held in method in this case
Step 38 is jumped in row again.
Otherwise ("No"), i.e., if the detection executed in step 42 the result is that negative, and correspondingly level error
Δ L is no more than threshold value, then in step 44, signal processor 10 reciprocally adjusts adjacent frequency band F3' and F4' amplification
The factor so that reach more than in step 42 detection previously given threshold value increased level error Δ L ' (Δ L '=|
L1‘–L2‘|；Referring to Fig. 4 and Fig. 5).Optionally, herein according to previously given standard come the increase of limit levels difference.Therefore,
In this case, the increase of level error is carried out so that be no more than previously given maximum value.It is covered in the present invention various
In the case of, the amplification factor before or after change can also have the value less than 1, and therefore cause input signal Ek
Frequency selective attenuation, even if this is atypical for traditional hearing aid.
Signal processor 10 especially calculates this change of amplification factor herein so that adjacent frequency band F3' and F4' in
Horizontal increase and horizontal reduction mutually compensates, that is to say, that frequency band F3' and F4' in dominant frequency f0Or f0' the tune at place
Signal level L after whole1' and L2' (or on average value) corresponds to the corresponding horizontal L before horizontal adjustment in summation1With
L2(L1‘+L2'=L1+L2).Different from simply asking summation or average value, one in the method extends embodiment
In, it is also contemplated that the amplitude frequency response of associated frequency band.
Then, signal processor 10 jumps to step 38 again in method execution.
Pass through the adjacent frequency band F carried out in step 443' and F4' in amplification factor reciprocal change, it is real
The advantage tone that can be heard with approximately equal intensity in output signal A is showed, just as there is no to carry out the level in step 44
Adjustment is the same.According to dominant frequency f0It is stronger in which of signal component NF and HF, herein with unmigrated frequency f0
Or with the frequency f of offset0' hear advantage tone.However, here, due to increased level error Δ L ', frequency f0And f0' it
Between the interference puppet sound of beat form be inhibited.
Within the scope of the invention, the implementation of many replacements of the method is possible.Such as frequency distortion (step
34) it can also be carried out in the other positions of this method flow, such as after level changes (step 42).In addition, in the present invention
In the range of, various other signal processing steps can also be executed between step 30 and 38, be especially used for noise suppressed
To input signal EkFrequency selectivity amplification and etc..
Illustrate adjacent frequency band F according to the present invention again according to Fig. 4 and Fig. 53' and F4' in the horizontal effect changed.
Here, from the clearer of the two figures, direction and dominant frequency f that level changes0Spectrum position it is related.If
According to the diagram in Fig. 4, dominant frequency is predominantly located at (f in high frequency component signal HF0>fg), then improve adjacent high frequency band
F4' signal level L2, and reduce adjacent low-frequency band F3' signal level L1, to increase level error Δ L.Conversely, such as
Fruit dominant frequency f0It is predominantly located at (f in low-frequency signal components NF0<fg), then improve adjacent low-frequency band F3' signal level
L1, and reduce adjacent high frequency band F4' signal level L2。
According to previously described embodiment, the present invention becomes especially clear.But the present invention is still not limited to these realities
Apply example.On the contrary, the numerous other embodiments of the present invention can be exported from the foregoing description.
Reference numerals list
2 hearing aids
4 input converters
6 subtracters
8 (analysis) filter groups
10 signal processors
12 (synthesis) filter groups
14 output translators
16 (electricity) feedback paths
18 (adaptive) filters
20 (acoustics) feedback paths
22 analogue-to-digital converters
24 digital-analog convertors
30 steps
32 steps
34 steps
36 steps
38 steps
40 steps
42 steps
44 steps
Δ L level errors
Δ L ' (increased) level error
F frequencies
f0(advantage) frequency
f0' (advantage after offset) frequency
fgEdge frequency
G signal gains
A output signals
Ei(initial) input signal
Ek(having been compensated for) input signal
FjFrequency band (j=1,2 ..., 6))
Fj' frequency band (j=1,2 ..., 6))
HF (high frequency) signal component
K thermal compensation signals
L1Signal level
L2Signal level
L1' signal level
L2' signal level
NF (low frequency) signal component
P (treated) signal
R feedback signals
RHFringe region
RLFringe region
Sa(output) voice signal
Si(input) voice signal
The overlapping regions U
Claims (12)
1. one kind when running hearing devices (2) for especially making as input signal (E existing for audio signalk) frequency mistake
Genuine method,
Wherein, by input signal (Ek) it is divided into low-frequency signal components (NF) and high frequency component signal (HF), wherein the two signals
Component (NF, HF) is in edge frequency (fg) at mutual interfaces,
Wherein, make high frequency component signal (HF) frequency distortion, and
Wherein, the high frequency component signal (HF) of low-frequency signal components (NF) and frequency distortion is superimposed, to form output signal
(A),
It is characterized in that, at least for including edge frequency (fg) high frequency component signal (HF) and/or low-frequency signal components (NF)
Fringe region (RH, RL), change associated amplification factor so that the signal level (L of low-frequency signal components (NF)1) and frequency
Signal level (the L of the high frequency component signal (HF) of rate distortion2) between level error (Δ L) increase.
2. according to the method described in claim 1,
Wherein, by filter group (8) by input signal (Ek) it is divided into multiple frequency band (Fj；J=1-6 therein multiple in)
Low-frequency band (F1-F3) conduction low-frequency signal components (NF), and wherein subsequent multiple high frequency band (F4-F6) Guiding high frequency
Signal component (HF),
Wherein, the fringe region (R of high frequency component signal (HF)H) by with low-frequency band (F1-F3) have a common boundary high frequency band (F4-
F6) subset formed, and/or wherein, the fringe region (R of low-frequency signal components (NF)L) by with high frequency band (F4-F6) have a common boundary
Low-frequency band (F1-F3) subset formed, and
Wherein, only with fringe region (RL, RH) associated frequency band (F3, F4) in change amplification factor.
3. method according to claim 1 or 2,
Wherein, the fringe region (R at least for the first signal component in two signal components (NF, HF)L, RH) increase amplification
The factor, and wherein, at least for second signal component (HF；NF fringe region (R)H；RL) reduce amplification factor.
4. according to the method described in claim 3,
Wherein, reduce second signal component (HF；NF amplification factor) so that the amplification of the first signal component of compensation (NF, HF)
The increase of the factor.
5. method according to claim 1 to 4,
Wherein it is determined that the input signal in the overlapping region (U) of characterization high frequency component signal (HF) and low-frequency signal components (NF)
(EK) tone feature characteristic parameter, and wherein, the change of amplification factor is carried out according to the characteristic parameter.
6. according to the method described in claim 5,
Wherein, the characteristic parameter for characterizing the feature of tone passes through the input signal (E in overlapping region (U)k) auto-correlation come it is true
It is fixed.
7. a kind of hearing devices (2),
There is frequency divider (8), be configured for, by input signal (Ek) it is divided into low-frequency signal components (NF) and high-frequency signal point
It measures (HF), wherein the two signal components (NF, HF) are in edge frequency (fg) at mutual interfaces,
There is signal processor (10), be configured for, make high frequency component signal (HF) frequency distortion, and
There is synthesizer (12), be configured for, by the high frequency component signal of low-frequency signal components (NF) and frequency distortion
(HF) it is superimposed, to form output signal (A),
It is characterized in that,
Signal processor (10) is configured for, at least for including edge frequency (fg) high frequency component signal (HF) and/or low
Fringe region (the R of frequency signal component (NF)H, RL), change amplification factor so that the signal level of low-frequency signal components (NF)
(L1) and frequency distortion high frequency component signal (HF) signal level (L2) between level error (Δ L) increase.
8. hearing devices (2) according to claim 7,
Wherein, frequency divider is formed by filter group (8), and the filter group is configurable for, by input signal (Ek) divide
To multiple frequency band (Fj；J=1-6 in), multiple low-frequency band (F therein1-F3) conduction low-frequency signal components (NF), and wherein
Subsequent multiple high frequency band (F4-F6) Guiding high frequency signal component (HF), wherein with low-frequency band (F1-F3) have a common boundary high frequency
Frequency band (F4-F6) subset formed high frequency component signal (HF) fringe region (RH), and/or wherein, with high frequency band (F4-F6)
Low-frequency band (the F of boundary1-F3) subset formed low-frequency signal components (NF) fringe region (RL), also,
Wherein, signal processor (10) is configured for, only with fringe region (RH, RL) associated frequency band (F3, F4) in change
Become amplification factor.
9. hearing devices (2) according to claim 7 or 8,
Wherein, signal processor (10) is configured for, at least for the first signal component in two signal components (HF, NF)
Fringe region (RH, RL) increase amplification factor, and at least for second signal component (NF；HF fringe region (R)L, RH)
Reduce amplification factor.
10. hearing devices (2) according to claim 9,
Wherein, signal processor (10) is configured for, and reduces second signal component (NF；HF amplification factor) so that compensation
The increase of the amplification factor of first signal component (HF, NF).
11. hearing devices (2) according to any one of claims 7 to 10,
Wherein, signal processor (10) is configured for, and determines characterization high frequency component signal (HF) and low-frequency signal components (NF)
Overlapping region (U) in input signal (Ek) tone feature characteristic parameter, and and if only if characteristic parameter meet it is advance
When given standard, it is amplified the change of the factor.
12. hearing devices (2) according to claim 11,
Wherein, signal processor (10) is configured for, and passes through the input signal (E in overlapping region (U)k) auto-correlation come it is true
Surely the characteristic parameter of the feature of characterization tone.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| DE102017203630.3 | 2017-03-06 | ||
| DE102017203630.3A DE102017203630B3 (en) | 2017-03-06 | 2017-03-06 | Method for frequency distortion of an audio signal and hearing device operating according to this method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108540913A true CN108540913A (en) | 2018-09-14 |
| CN108540913B CN108540913B (en) | 2020-08-11 |
Family
ID=60972136
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810181968.7A Active CN108540913B (en) | 2017-03-06 | 2018-03-06 | Method for frequency distorting an audio signal and hearing device operating according to the method |
Country Status (7)
| Country | Link |
|---|---|
| US (1) | US10674283B2 (en) |
| EP (1) | EP3373599B1 (en) |
| JP (1) | JP6622829B2 (en) |
| CN (1) | CN108540913B (en) |
| AU (1) | AU2018200907A1 (en) |
| DE (1) | DE102017203630B3 (en) |
| DK (1) | DK3373599T3 (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10483931B2 (en) * | 2017-03-23 | 2019-11-19 | Yamaha Corporation | Audio device, speaker device, and audio signal processing method |
| DE102021205251A1 (en) * | 2021-05-21 | 2022-11-24 | Sivantos Pte. Ltd. | Method and device for frequency-selective processing of an audio signal with low latency |
| US20230326474A1 (en) * | 2022-04-06 | 2023-10-12 | Analog Devices International Unlimited Company | Audio signal processing method and system for noise mitigation of a voice signal measured by a bone conduction sensor, a feedback sensor and a feedforward sensor |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6115478A (en) * | 1997-04-16 | 2000-09-05 | Dspfactory Ltd. | Apparatus for and method of programming a digital hearing aid |
| US6236731B1 (en) * | 1997-04-16 | 2001-05-22 | Dspfactory Ltd. | Filterbank structure and method for filtering and separating an information signal into different bands, particularly for audio signal in hearing aids |
| US6240192B1 (en) * | 1997-04-16 | 2001-05-29 | Dspfactory Ltd. | Apparatus for and method of filtering in an digital hearing aid, including an application specific integrated circuit and a programmable digital signal processor |
| US6738486B2 (en) * | 2000-09-25 | 2004-05-18 | Widex A/S | Hearing aid |
| CN1607725A (en) * | 2003-10-15 | 2005-04-20 | 罗姆股份有限公司 | Sound quality enhancement circuit for audio signals and audio amplifier circuit using the same |
| US20100272289A1 (en) * | 2009-04-24 | 2010-10-28 | Siemens Medical Instruments Pte. Ltd. | Method for operating a hearing apparatus and hearing apparatus with a frequency separating filter |
| CN101897118A (en) * | 2007-12-11 | 2010-11-24 | Nxp股份有限公司 | Prevention of audio signal clipping |
Family Cites Families (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2563719B2 (en) | 1992-03-11 | 1996-12-18 | 技術研究組合医療福祉機器研究所 | Audio processing equipment and hearing aids |
| JPH0956000A (en) | 1995-08-11 | 1997-02-25 | Nippon Telegr & Teleph Corp <Ntt> | Hearing aid |
| EP1016318A1 (en) | 1998-07-02 | 2000-07-05 | Beltone Netherlands B.V. | Hearing instrument |
| EP2091266B1 (en) * | 2008-02-13 | 2012-06-27 | Oticon A/S | Hearing device and use of a hearing aid device |
| DK2217007T3 (en) * | 2009-02-06 | 2014-08-18 | Oticon As | Hearing aid with adaptive feedback suppression |
| DE102009021310B4 (en) * | 2009-05-14 | 2011-02-24 | Siemens Medical Instruments Pte. Ltd. | Binaural hearing apparatus and method for operating a binaural hearing apparatus with frequency distortion |
| CN102264022B (en) | 2010-04-08 | 2014-03-12 | Gn瑞声达公司 | Stability improvements in hearing aids |
| DE102010026884B4 (en) * | 2010-07-12 | 2013-11-07 | Siemens Medical Instruments Pte. Ltd. | Method for operating a hearing device with two-stage transformation |
| US8855342B2 (en) * | 2010-09-29 | 2014-10-07 | Siemens Medical Instruments Pte. Ltd. | Hearing aid device for frequency compression |
| JP2012175453A (en) | 2011-02-22 | 2012-09-10 | Sony Corp | Speech processing device, speech processing method and program |
| DK3122072T3 (en) * | 2011-03-24 | 2020-11-09 | Oticon As | AUDIO PROCESSING DEVICE, SYSTEM, USE AND PROCEDURE |
| DK2988529T3 (en) | 2014-08-20 | 2020-02-24 | Sivantos Pte Ltd | ADAPTIVE DISTRIBUTION FREQUENCY IN HEARING AID DEVICES |
| DE102015216822B4 (en) * | 2015-09-02 | 2017-07-06 | Sivantos Pte. Ltd. | A method of suppressing feedback in a hearing aid |
-
2017
- 2017-03-06 DE DE102017203630.3A patent/DE102017203630B3/en not_active Expired - Fee Related
-
2018
- 2018-01-15 DK DK18151664T patent/DK3373599T3/en active
- 2018-01-15 EP EP18151664.2A patent/EP3373599B1/en active Active
- 2018-02-07 AU AU2018200907A patent/AU2018200907A1/en not_active Abandoned
- 2018-02-09 US US15/893,097 patent/US10674283B2/en active Active
- 2018-03-06 JP JP2018039291A patent/JP6622829B2/en active Active
- 2018-03-06 CN CN201810181968.7A patent/CN108540913B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6115478A (en) * | 1997-04-16 | 2000-09-05 | Dspfactory Ltd. | Apparatus for and method of programming a digital hearing aid |
| US6236731B1 (en) * | 1997-04-16 | 2001-05-22 | Dspfactory Ltd. | Filterbank structure and method for filtering and separating an information signal into different bands, particularly for audio signal in hearing aids |
| US6240192B1 (en) * | 1997-04-16 | 2001-05-29 | Dspfactory Ltd. | Apparatus for and method of filtering in an digital hearing aid, including an application specific integrated circuit and a programmable digital signal processor |
| US6738486B2 (en) * | 2000-09-25 | 2004-05-18 | Widex A/S | Hearing aid |
| CN1607725A (en) * | 2003-10-15 | 2005-04-20 | 罗姆股份有限公司 | Sound quality enhancement circuit for audio signals and audio amplifier circuit using the same |
| CN101897118A (en) * | 2007-12-11 | 2010-11-24 | Nxp股份有限公司 | Prevention of audio signal clipping |
| US20100272289A1 (en) * | 2009-04-24 | 2010-10-28 | Siemens Medical Instruments Pte. Ltd. | Method for operating a hearing apparatus and hearing apparatus with a frequency separating filter |
Also Published As
| Publication number | Publication date |
|---|---|
| JP6622829B2 (en) | 2019-12-18 |
| DE102017203630B3 (en) | 2018-04-26 |
| EP3373599A1 (en) | 2018-09-12 |
| CN108540913B (en) | 2020-08-11 |
| AU2018200907A1 (en) | 2018-09-20 |
| EP3373599B1 (en) | 2019-08-21 |
| US10674283B2 (en) | 2020-06-02 |
| DK3373599T3 (en) | 2019-11-25 |
| JP2018148561A (en) | 2018-09-20 |
| US20180255405A1 (en) | 2018-09-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP1742509B1 (en) | A system and method for eliminating feedback and noise in a hearing device | |
| EP1538868B1 (en) | Audio amplification apparatus | |
| US8976988B2 (en) | Audio processing device, system, use and method | |
| EP2551845B1 (en) | Noise reducing sound reproduction | |
| US7483831B2 (en) | Methods and apparatus for maximizing speech intelligibility in quiet or noisy backgrounds | |
| US9325285B2 (en) | Method of reducing un-correlated noise in an audio processing device | |
| US8693717B2 (en) | Method for compensating for an interference sound in a hearing apparatus, hearing apparatus, and method for adjusting a hearing apparatus | |
| CN108540913A (en) | Make the method that audio signal frequency is distorted and the hearing devices to work according to this method | |
| US10142742B2 (en) | Audio systems, devices, and methods | |
| US9373317B2 (en) | Headset and headphone | |
| KR101405847B1 (en) | Signal Processing Structure for Improving Audio Quality of A Car Audio System | |
| CN115398934A (en) | Method, device, earphone and computer program for actively suppressing occlusion effect when reproducing audio signals | |
| DK2437521T4 (en) | Method for frequency compression with harmonic correction and corresponding apparatus | |
| CN110505558B (en) | Method for operating a hearing aid | |
| KR102403996B1 (en) | Channel area type of hearing aid, fitting method using channel area type, and digital hearing aid fitting thereof | |
| US20230007408A1 (en) | Hearing instrument and method for directional signal processing of signals in a microphone array | |
| US20230396939A1 (en) | Method of suppressing undesired noise in a hearing aid | |
| JP4703771B1 (en) | Acoustic signal correcting apparatus and acoustic signal correcting method | |
| JP2021157134A (en) | Signal processing method, signal processing device and hearing device | |
| CN117376798A (en) | Directional signal processing method for hearing devices |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |